MantisBT - MEGA
View Issue Details
IDProjectCategoryView StatusDate SubmittedLast Update
0000915MEGACalculation Errorpublic2018-09-07 08:222018-09-12 10:47
Reporterguest 
Assigned Togstecher 
PrioritynormalSeverityminorReproducibilityhave not tried
StatusresolvedResolutionfixed 
PlatformPCOSWindowsOS Version7
Product Version 
Target VersionFixed in Version 
First NameKathrin
Last NameSeibt
Emailkathrin.seibt@tu-dresden.de
Confirm Emailkathrin.seibt@tu-dresden.de
Summary0000915: Distance Matrix Calculation - Bug?
DescriptionHallo MEGA support,

I am trying to calculate a distance matrix for some non-coding nucleotide sequencesusing MEGA7 as MEGA-CC on Windows 7. I wanted to try different parameters to inspect the results. Surprisingly, I got exactly the same distance matrices when using p-distance + uniform rates vs. Tamura 3-parameter model + Gamma distributed. All 276 values in the matrix were identical, but the stated settings differed. Is this a bug or is it indeed identical?

Thanks in advance!
Kathrin
Steps To Reproducemegacc -d Seq.fasta -f Fasta -a b-pairw-distance-estimation_nt_pdist_pairwise-del.mao
Additional Information
TagsNo tags attached.
Relationships
Attached Filesrar MEGA7-DistCalc.rar (3,296) 2018-09-07 08:22
https://megasoftware.net/mantis_bt/
Issue History
Date ModifiedUsernameFieldChange
2018-09-07 08:22guestNew Issue
2018-09-07 08:22guestFile Added: MEGA7-DistCalc.rar
2018-09-12 10:23gstecherFile Deleted: bugreport.txt
2018-09-12 10:47gstecherNote Added: 0004109
2018-09-12 10:47gstecherStatusnew => resolved
2018-09-12 10:47gstecherResolutionopen => fixed
2018-09-12 10:47gstecherAssigned To => gstecher

Notes
(0000842)
paul   
1969-12-31 17:33   
Request detail information and data from user.
(0001068)
paul   
1969-12-31 17:33   
Not a bug, time to load data depends on the power of computer. For a big dataset, it may take a while to load. Close the report.
(0004109)
gstecher   
2018-09-12 10:47   
Hi Kathrin,

I am writing in response to the bug report you recently submitted regarding the MEGA software. Thanks for the feedback. No, the results you obtained should not have been the same when using different parameters for distance estimation. It seems that there is a bug in the MEGA-Proto application that generates the .mao file when using a non-English (US) locale. The d-pairw-distance-estimation_nt_Tamura3_pairwise-del_Gamma.mao file has the gamma parameter value set to 1,00 but megacc expects it should be 1.00 (decimal separator should be the '.' character, not the ',' character). If you open that file in a text editor and change the line

Gamma Parameter = 1,00

to

Gamma Parameter = 1.00

then you should get the correct result. This issue should not be a problem in the current version of MEGA which can be downloaded from:

https://www.megasoftware.net/releases/MEGA_X_CC_10.0.4_win64_setup.exe [^]


--
Best regards,

Glen Stecher
Institute for Genomics and Evolutionary Medicine
igem.temple.edu